Apr 2021 – Nov 2021 · 7 mos · Bengaluru, Karnataka, India

• Developed ETL pipelines using Airflow
• Deployed Covid Resources Rasa chatbot on GCP
• Set up Airflow cluster on EC2 following multi node architecture
• Developed ETL to mask the user id's (PII) using cryptographic algorithms and orchestrated it using Airflow

Oct 2018 – Apr 2021 · 2 yrs 6 mos

• Techniques: Regression, Classification, Dimension reduction, Model Selection, Natural Language Processing, Neural Networks.
• Tools: Python, Shell Script, Tableau, HIVE, HS2, Kibana, Tez, YARN, Ambari, Jenkins
• Developed an engine which identifies the most important topics or subjects from all the cluttered Service Now incidents, problem tickets and change tickets so as to organize, understand and summarize large collections of textual information (Topic Modelling)
• Developed Classification engine to detect and mask Personally Identifiable Information (PII) from all the sources to protect the personal data of users
• Developed Recommendation engine (SVD, NMF) to allow users to discover similar reports, creating a delightful user experience while driving incremental value
• Developed an algorithm to detect acute intracranial hemorrhage and its sub-types using MRI scans (CNN)
• Investigated, proposed, developed and refined new analytic capabilities for deployment in the business
• Built algorithms, tools and custom solutions for use by business analysts and forecasters
• Developed frameworks that integrate Big Data and advanced analytics to make business decisions
• Involved in all stages of data science product development from definition, design, development and delivery
• Worked with the application development teams during application design and development for highly complex and critical data projects
• Developed data processing pipelines using tools and frameworks in the Hadoop ecosystem
• Developed dashboards using data visualization tools to facilitate data analysis

Jul 2017 – Oct 2018 · 1 yr 3 mos

• Tools: HVR, Talend, Postgres, Tableau
• Worked with Vendors to tune performance of Data replication mechanism using HVR
• Proposed and implemented ETL pipelines to automate ingestion of structured data
• Investigated & Proposed performance tuning of GP queries used to integrate data into FDL
• Develop framework, metrics & reporting to ensure progress can be measured, evaluated & continually improved
• Developed dashboards using data visualization tools to facilitate data analysis

Jan 2017 – Jun 2017 · 5 mos

• Tools: HVR, Oozie
• Developed data replicating pipelines to ingest Legacy data to FDL Staging area using HVR
• Developed workflows using Oozie to schedule Hadoop jobs which replicate FDL data to Consumption layer